INDEX

Explanations

commas indicating pauses or breaks in sentences

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.85

ãĤ¢ãĥ«

-0.78

Optional

-0.77

Otherwise

-0.77

Appearance

-0.74

ãĤ¨ãĥ«

-0.71

anus

-0.70

 Otherwise

-0.69

afety

-0.69

AB

-0.67

POSITIVE LOGITS

 however

1.07

 successive

0.88

 moreover

0.80

 though

0.74

 whenever

0.73

 culminating

0.70

 dozens

0.70

 countless

0.69

 meanwhile

0.68

 researchers

0.68

Activations Density 0.120%

commas indicating pauses or breaks in sentences

No Comments

No Known Activations